Project Summary Broadly neutralizing antibodies (bNAbs) recognize relatively conserved epitopes on the envelope glycoprotein (Env) of human immunodeficiency virus type-1 (HIV-1). High-resolution structures of a cleaved, soluble BG505 SOSIP.664 gp140 trimer have provided a rational basis for HIV-1 trimer vaccine design. Although the SOSIP mutations have proven very successful in stabilizing Env trimer, the cause of HIV trimer metastability remains unclear. Thus, there is still room for the evaluation of alternative platforms for trimer stabilization. Furthermore, there is increasing appreciation for the advantages of virus-like particle (VLP) vaccines and the importance of spike density and spacing in bNAb elicitation. However, such particulate display of antigens has yet to be rigorously explored for an HIV-1 vaccine. In this proposal, we will address the challenges of HIV-1 trimer stabilization and particulate vaccines by combining rational protein design, nanoparticle engineering, animal immunization, serum binding and neutralization, single B-cell and B-cell repertoire analyses. In Aim 1, we will develop and optimize a novel gp140 trimer platform termed uncleaved, prefusion-optimized (UFO) trimer. We hypothesize that both the HR1 N-terminus (between ?6 and ?7) and the cleavage site are key contributors to HIV-1 trimer metastability, and that the rational redesign of these two regions will render a general strategy for trimer stabilization. We tested ~20 designs based on BG505 and observed substantially improved trimer yield and purity. Based on BG505 and other HIV-1 strains, we will characterize the structure and antigenicity of UFO trimers to investigate trimer metastability and their potential as trimer immunogens. In Aim 2, we will develop particulate HIV-1 immunogens. We hypothesize that self-assembling nanoparticles with 3-fold vertices can be used to display trimeric HIV-1 antigens as multivalent vaccines. Based on ferritin (12.2 nm), we have designed V1V2, gp120, and gp140 nanoparticles, confirmed their assembly with electron microscopy, and assessed their antigenicity using a panel of antibodies. Based on dihydrolipoyl acyltransferase (E2p) from Geobacillus stearothermophilus (23.2 nm), we aim to design VLP-like vaccine candidates to elicit robust B cell responses. In Aim 3, we will assess immunogenicity and B cell responses of UFO trimers and particles in both rabbits and nonhuman primates (NHPs). We hypothesize that the UFO trimer based on a tier-2 isolate will elicit autologous NAb responses perhaps with some degree of breadth, while the particles with 8 to 20 native-like trimer spikes will elicit more potent NAb responses. We further hypothesize that a heterologous regimen with immunogens from different clades will increase the breadth of elicited NAb responses. We will assess the immunogenicity and regimens in rabbits and select the best trimer and particle immunogens with the optimal regimen for further evaluation in NHPs. We will use antibody NGS to derive quantitative B cell repertoire profiles and antibody lineage patterns, in addition to serum binding and neutralization profile. With iterative immunogen design and immunization, we set out to assess and extend the potential of protein-based HIV-1 trimer vaccines.